INTERCHANGEABLE PARTS, of in dustrial or military implements, are the key to all the vast expansion of modern manufactur ing. Without these, only a small fraction of it, and consequently of new population human or animal, agriculture, engineering and industrial civilization in general, could have come into being. A glance at its implications will show that this seemingly extravagant proposition is sober fact. ..
The principle:,of interchangeability affects production in two ways: through the original manufacturing,, and through repairing. In manufacture it makes possible the turning out, in separate departments, in vast masses at full speed, in exact shape for immediate assembling, of all the individual parts used in great num bers in complicated machines, from the screws of a watch to the rivets of. a giant steamship. Without this, the different parts would re quire an army of fitters to make them work with each other, and indefinitely more time for assembling, hence a much diminished produc tion and higher expense. In repair, if every re placed part of a watch or sewing-machine, a gun or a reaper, had to be individually filed or hammered or reforged into adjustment, the delay and expense would greatly decrease the articles' use and the work done by their means; and for military service, repair would involve so much remanufacturing on the field as to make modern warfare quite impracticable. Now, reflecting that the multiplication of men is conditioned by that of goods and employ ment, of food by farms and transportation, of the latter by engineering, and so on, the conclusion above is easily substantiated.
Owing to mechanical niceties to be ex plained later, the principle did not attain much close accuracy in use till toward the middle of the last century; and as might be expected, was first developed in making firearms, where masses of lives were involved and the highest of stakes to be won. The Springfield, Mass., armory was the first to bring it to a high grade of embodiment. Somewhat later, the Waltham Watch Company applied it, under much more difficult conditions, to the minute, sometimes almost microscopic, parts of its watches. The constant effort of all large implement manu facturing for many years has been toward perfecting its application; and its success largely gauges the increased use of machinery in place of men, and of mechanical utilities.
The first requisite is obviously that each part shall come within very small limits of variation, neither stick nor "shuck° ; the one stopping or slacking movement and increasing frictional wear and heat, the other making movement ineffective and irregular and break ages likely. Or if, say, a fuse, its cavity must hold pretty closely the same amount of ex plosive to determine time and goal. Hence the standards of accuracy and craftsmanship must be kept very high; in the lone run it is most economical if worth attempting at all. Any saving on expense by slighting severe require ments is more than sunk on increased cost in assembling, to say nothing of rejected work and impaired reputation.
While every step must help to ensure this uniformity, it must be frankly accepted that there is a certainty of some error in each; and the proper course is not to attempt eliminating it wholly— an impossioie task, and beyond a certain limit costing more than it is worth,— but to determine the limits up to which it is neglectible, and hold it within them. Con stant accurate measurement for those limits is therefore the basic need; and the limits them selves depend on the class of work and close ness of adjustment required. Thus, if a piece is to be one inch in a given dimension, with an ordinary steel scale we are sure of it within about 1/100 inch; with a micrometer, about 1/4000; on a measuring machine, about 1/100,000. The latter is not available in most small establishments; and when the standard of accuracy must be held high, the best means of guarding against discrepancies from inde pendent measurements is by the establishment of a model. All measurements are then com parative instead of direct, and correct within much smaller limits of error. Ifthis be neglected, its lack will be a constant hindrance to the high standard requisite in all the better classes of implements; and no other system so economical and reliable has yet been devised within the general reach. An example is the 36-inch bronze bar in government archives as a legal standard of length. Drawings of each part for general shop reference should also be provided, with all dimensions and tolerances clearly shown.